**** BEGIN LOGGING AT Wed Jul 02 03:00:00 2014
Jul 02 05:37:30 <vvu> morning!
Jul 02 09:40:39 <Abhishek_> Hey vvu
Jul 02 10:03:29 <vvu> hello Abhishek_
Jul 02 11:41:57 <karki_> interesting
Jul 02 11:43:10 <vvu> the silence around here ?
Jul 02 13:27:31 <karki_> vvu : yes
Jul 02 13:29:50 <rseethamraju> alexanderhiam: VoltVisionFrench: I figured out whats going wrong. It only gives the right temp value when I reset the bbb after each time I connect to the sensor.The getTemp() part works. The other parts don’t. I need to be able to reset the sensor programatically. The datasheet says that to reset, I have to send a Logic 1 to the sensor’s DIN (thats the BBB’s DOUT) for 32 clock cycles. How do I count the no of cycles? Should I count in a
Jul 02 13:29:50 <rseethamraju> for loop the delay from the default max frequency thats used?
Jul 02 13:33:20 <Abhishek_> rseethamraju: is it a 1wire?
Jul 02 13:33:31 <rseethamraju> 4 wire SPI
Jul 02 13:35:00 <_av500_> vvu: I will be in a shitty paris suburb train at 6pm
Jul 02 13:35:09 <_av500_> can you cover for me?
Jul 02 13:35:13 <vvu> yep
Jul 02 13:35:26 <_av500_> thx
Jul 02 13:35:30 <vvu> np
Jul 02 14:46:11 <Abhishek_> gm jkridner
Jul 02 14:46:46 <_av500_> jkridner: hi
Jul 02 14:46:50 <_av500_> jkridner: im stuck in paris
Jul 02 14:46:51 <jkridner> hi Abhishek_, _av500_
Jul 02 14:47:03 <_av500_> wont make it at the meeting, suburb trains have no wifi
Jul 02 14:47:10 <jkridner> _av500_: k
Jul 02 14:48:33 * jkridner checks status reports ahead of the meeting.
Jul 02 15:08:39 * vmayoral|home will be back for meeting time ;)
Jul 02 15:12:35 <Abhishek_> vmayoral: Let me know if BeagleLogic can help you troubleshoot your sensor
Jul 02 15:14:34 <alexanderhiam> rseethamraju: just sent you an email
Jul 02 15:15:04 <rseethamraju> yes I saw
Jul 02 15:15:09 <alexanderhiam> I wonder if that's just how continuous mode works
Jul 02 15:15:25 <rseethamraju> I’ll use the 1 SPS mode
Jul 02 15:16:01 <alexanderhiam> wait, isn't there also an on-demand measurement mode?
Jul 02 15:16:06 <rseethamraju> so I have to send a command byte and read it simultaneoudly
Jul 02 15:16:24 <rseethamraju> I’ll have to change spi’s transfer() too
Jul 02 15:17:16 <rseethamraju> because right now it only reads one byte. It gas to read atleat 2 bytes for the temp sensor.
Jul 02 15:17:16 <alexanderhiam> oh right, I was thinking one-shot mode, forgot it went to shutdown after the first conversion
Jul 02 15:18:04 <alexanderhiam> oh wait, I think I got it
Jul 02 15:18:16 <alexanderhiam> in the continuous mode description:
Jul 02 15:18:17 <rseethamraju> ya I tries that. So in the one-shot mode it has to write and read on the next clock cycle which is a problem
Jul 02 15:18:29 <alexanderhiam> "B
Jul 02 15:18:29 <alexanderhiam> y
Jul 02 15:18:29 <alexanderhiam> sending 16
Jul 02 15:18:29 <alexanderhiam> S
Jul 02 15:18:29 <alexanderhiam> CLK clocks to the ADT7310
Jul 02 15:18:30 <alexanderhiam> ,
Jul 02 15:18:32 <alexanderhiam> the c
Jul 02 15:18:34 <alexanderhiam> ontents of the temperature
Jul 02 15:18:36 <alexanderhiam> value
Jul 02 15:18:38 <alexanderhiam> register
Jul 02 15:18:42 <alexanderhiam> are
Jul 02 15:18:44 <alexanderhiam> output
Jul 02 15:18:46 <alexanderhiam> onto the DOUT pin."
Jul 02 15:18:48 <alexanderhiam> whoa... copy and paste weirdness
Jul 02 15:18:52 <rseethamraju> pg no?
Jul 02 15:19:13 <alexanderhiam> but that explains the behavious, it's ignoring the data you send it completely, just looking at the clock
Jul 02 15:19:51 <alexanderhiam> then it says: "To exit the continuous read mode, the Command Byte 01010000 (0x50) must be written to the ADT7310"
Jul 02 15:21:08 <alexanderhiam> so you only write the continuous mode command the first time, then just read two bytes to get the value each time, and if you want to change anything else you have to stop continuous mode
Jul 02 15:21:41 <rseethamraju> but its reading the temp out of thge temp register fine
Jul 02 15:22:00 <rseethamraju> can I show you now?
Jul 02 15:22:08 <alexanderhiam> right, but that's all it can do when in continuous mode
Jul 02 15:22:43 <alexanderhiam> sure, hangout?
Jul 02 15:23:19 <rseethamraju> ok
Jul 02 15:23:39 <jkridner> only 3 reports so far? BARD, bone101 and beaglepilot?
Jul 02 15:24:24 <jkridner> karki: no report on pruspeak yet?
Jul 02 15:24:43 <jkridner> alexanderhiam: do you know if praveen is joining?
Jul 02 15:25:38 <jkridner> rseethamraju, Abhishek_: thanks for getting your reports in ahead of the meeting.
Jul 02 15:25:45 <alexanderhiam> jkridner: I do not know
Jul 02 15:27:05 <jkridner> rseethamraju: oops, I didn't mean to say I saw your report.
Jul 02 15:27:26 <jkridner> vvu: do you know if praveen is joining?
Jul 02 15:27:47 <Abhishek_> jkridner: done
Jul 02 15:28:38 <jkridner> Thanks Abhishek_
Jul 02 15:29:59 <jkridner> ok, now I have reports from Abhishek_, praveen, victor and Diego... still needed for rseethamraju and karki
Jul 02 15:30:46 <karki_> jkridner : in a few minutes :)
Jul 02 15:40:09 <Abhishek_> jkridner: I had told you that I had captured a 1 GB dump (= 100 seconds) using the driver [not these bindings] very well, and it wasn't *that* bad even with that size.
Jul 02 15:41:51 <jkridner> I meant when using sigrok-cli
Jul 02 15:42:03 <jkridner> new bottleneck.
Jul 02 15:42:13 <ds2> wonder if sigrok-cli is memory intensive
Jul 02 15:42:33 <Abhishek_> They use zlib in the backend to compress, the CPU light was constantly on
Jul 02 15:42:43 <Abhishek_> did you see the load average in the screenshot?
Jul 02 15:43:41 <Abhishek_> jkridner: good issue to talk about with biot in tomorrows's hangout
Jul 02 15:43:50 <Abhishek_> *tomorrow's
Jul 02 15:44:03 <jkridner> yup
Jul 02 15:45:23 <VoltVisionSteve> Hi all!  I have to leave @ 12:20 today...I am here, but I am working with an employee on another project, just ping me as necessary.  Thx!
Jul 02 15:45:48 <vvu> jkridner: he should
Jul 02 15:46:07 <jkridner> VoltVisionSteve: k
Jul 02 15:48:22 <Abhishek_> jkridner: the protocol decoders run on python and are very CPU-intensive, not much card activity but kept the CPU spinning pretty long, as I quoted it took me 800secs to decode a 6sec dump
Jul 02 15:49:03 <ds2> Abhishek_: I have seen tek scopes take around that time for the large dumps
Jul 02 15:49:10 <jkridner> Abhishek_: I'm just wondering if swapping becomes an issue over a certain size which might be the practical limit.
Jul 02 15:49:27 <jkridner> ds2: +1
Jul 02 15:50:29 <Abhishek_> I don't go too deep into it, I don't think any swapping takes place in the decoding I tested, the data is in RAM and the kernel buffers are mmap()'ed into userspace, so no bottleneck on the kernel side
Jul 02 15:50:31 <jkridner> 256MB dump is pretty large. Most debugging shouldn't need anything that large.
Jul 02 15:50:35 <DiegoTc> VoltVisionSteve: did you send the email to Oscar yesterday?
Jul 02 15:51:06 <ds2> 1-2M is probally more realistic but the lack of a trigger makes that difficult
Jul 02 15:51:27 <Abhishek_> ds2: sigrok supports soft triggers now
Jul 02 15:52:09 <VoltVisionSteve> DiegoTc, yes I did send the email to Oscar yesterday.... I also just sent one now again to all of us and I explained the urgency of everything... feel free to chime in the email thread.
Jul 02 15:53:07 <ds2> Abhishek_: I mean HW/PRU FW triggers
Jul 02 15:53:19 <VoltVisionSteve> DiegoTc, did you find Paola's advice useful?
Jul 02 15:53:22 <ds2> doing it in Sigrok would is slow
Jul 02 15:53:51 <rseethamraju> jkridner: done. Sorry I’m late!
Jul 02 15:53:53 <Abhishek_> I'll see how realistic soft triggers can be there, I'll be working on it
Jul 02 15:55:15 <DiegoTc> VoltVisionSteve: not really, she basically told me that new websites are done with parallex
Jul 02 15:55:28 <DiegoTc> and a link for combining colors
Jul 02 15:55:54 <DiegoTc> jkridner: http://screencast.com/t/bxY8cB4xR I change the layout the for the editor, and use the bonecard view
Jul 02 15:56:15 <DiegoTc> Still can't move the toolbar
Jul 02 15:58:00 <jkridner> which tool are you using again?
Jul 02 15:58:12 <jkridner> for the editor
Jul 02 15:58:17 <vmayoral> Greetings everyone!
Jul 02 15:58:19 <Abhishek_> Greetings vmayoral!
Jul 02 15:58:34 <DiegoTc> jkridner: Summernote https://github.com/HackerWins/summernote
Jul 02 15:58:41 <jkridner> hi vmayoral. thanks for getting your report in early enough to read before the meeting
Jul 02 15:59:04 <Abhishek_> vmayoral: could BeagleLogic be able to help you debugging your MPU6K :) ?
Jul 02 16:00:37 <ds2> vmayoral: how erratic are the accel readings?
Jul 02 16:01:08 <vmayoral> Abhishek_: mmm interesting idea. We can explore it
Jul 02 16:01:16 <jkridner> DiegoTc: hmmmm... summernote might support only a single div for both editor and toolbar. :(
Jul 02 16:01:32 <jkridner> let's start the meeting....
Jul 02 16:01:55 <jkridner> before we start going around the room checking in on the projects (reviewing the status reports), does anyone have any additional topics to discuss for the group?
Jul 02 16:02:02 <praveendath92> Hello all
Jul 02 16:02:51 <rseethamraju> yes. has anyone used gstreamer with webcams?
Jul 02 16:03:02 <vmayoral> ds2: according to tridge they are bad. I'm still quite new to be able to judge them just by looking at accels and gyros values. Im trying to plot them directly from the autopilot
Jul 02 16:03:32 <ds2> rseethamraju: long long time ago..what are you thinking of?
Jul 02 16:03:50 <vmayoral> ds2: i wonder what would happen if i launxh the copterwiththe current noisy values
Jul 02 16:03:53 <ds2> vmayoral: okay.... is this a static reading or dynamic?
Jul 02 16:04:24 <ds2> vmayoral: with or w/o the FIFO's?
Jul 02 16:04:34 <rseethamraju> ds2: I’m using the v4l2src element to get video put of the webcam but the pipeline only works for the Logitech C920
Jul 02 16:05:00 <rseethamraju> alexanderhiam: has a Logitech C270 that doesn’t work with the same pipeline
Jul 02 16:05:03 <vmayoral> ds2: with FiFOs. Static/dynamic? What do u mean?
Jul 02 16:05:14 <jkridner> BTW, I am recording attendance for vmayoral, praveendath92, rseethamraju, DiegoTc, karki_, Abhishek_
Jul 02 16:05:24 <karki_> :)
Jul 02 16:05:34 <rseethamraju> present!
Jul 02 16:05:43 <Abhishek_> SYN
Jul 02 16:06:01 <karki_> jkridner : I'll put my weekly status in 2 min. sorry for the delay.
Jul 02 16:06:10 <jkridner> rseethamraju: I think that is part of the update, but we'll go ahead and cover yours first, then DiegoTc because VoltVisionSteve needs to head out very soon.
Jul 02 16:06:39 <Abhishek_> hi cdsteinkuehler
Jul 02 16:06:43 <rseethamraju> oh its ok you can do DiegoTc’s
Jul 02 16:07:25 <vmayoral> ds2: the gyros are used as the metronome of the autopilot. Readings are fast enogh using current SPI
Jul 02 16:07:28 <DiegoTc> you decide jkridner
Jul 02 16:07:45 <vmayoral> ds2: loop running at 100/200 Hz
Jul 02 16:07:49 <ds2> vmayoral: is things moving or just sitting still
Jul 02 16:07:59 <vmayoral> Mmmm
Jul 02 16:08:13 <ds2> rseethamraju: you might have to send it through a color space conversion filter
Jul 02 16:08:14 <jkridner> rseethamraju: we'll let the GStreamer talk go in the background a bit, but let's let VoltVisionSteve chime in on DiegoTc's status.
Jul 02 16:08:22 <rseethamraju> ok
Jul 02 16:08:39 <vmayoral> ds2:you mean for the tests :). The board is still for now
Jul 02 16:08:41 <jkridner> DiegoTc: it seems you are blocked on approval of the editor view.
Jul 02 16:08:48 <DiegoTc> yes
Jul 02 16:08:53 <jkridner> DiegoTc: I'm not happy with the editor toolbar disrupting the view....
Jul 02 16:09:01 <jkridner> It seems that might be a need to drop summernote.
Jul 02 16:09:05 <jkridner> VoltVisionSteve: opinions?
Jul 02 16:09:18 <DiegoTc> I was using this editor before: https://github.com/DiegoTc/bootstrap-wysiwyg but having issues for having multiple editors on a same page
Jul 02 16:09:22 <ds2> vmayoral: then bad readings should be pretty easy to spot... when it is still, |(X, Y, Z)| = 9.8m/s^2
Jul 02 16:09:34 <ds2> vmayoral: anything else is a problem.
Jul 02 16:09:58 <DiegoTc> there was a way to fixed but never try it. I will see it and test it today.
Jul 02 16:10:11 <jkridner> DiegoTc: k.
Jul 02 16:10:12 <rseethamraju> ds2: you mean like ffmpegcolorspace?
Jul 02 16:10:32 * VoltVisionSteve looking for most recent links
Jul 02 16:10:59 <ds2> rseethamraju: something like that... I suspect the other cams are producing an unrecogonized/incompatible color space relative to your pipeline
Jul 02 16:11:05 <jkridner> DiegoTc, VoltVisionSteve: you could also do something like an editor below the card and simply duplicate the content in the card live. That would also be especially handy if it included the ability to edit the source rather than just WYSIWYG
Jul 02 16:11:36 <jkridner> DiegoTc: dynamic content update of source would require some syntax checking to avoid disrupting the page view.
Jul 02 16:11:42 <vmayoral> ds2: i don't quite follow you. I expect x-accel to be -9.8 but the rest (gyros and accels) should be close to 0 , shouldn't they?
Jul 02 16:12:03 <DiegoTc> yes, I was thinking on that
Jul 02 16:12:27 <vmayoral> ds2: the current problem is that the current measurements although in the expected range, oscillate too much (that's my belief)
Jul 02 16:12:30 <DiegoTc> I will go first with the editor, if no your  last comment. I will let you know
Jul 02 16:12:51 <Abhishek_> vmayoral: are you using filters with your data?
Jul 02 16:13:11 <ds2> vmayoral: if X is aligned with gravity, yes.
Jul 02 16:13:23 <vmayoral> For stabilization yes
Jul 02 16:13:28 <Abhishek_> I mean, which filter are you using, the Madgwick filter or Kalman filter?
Jul 02 16:13:33 <vmayoral> Ardupilot implemmenta several
Jul 02 16:13:44 <jkridner> DiegoTc: k. Please publish URL and let the group know you are looking for testing to get feedback quickly to move on to starting to make some tutorials.
Jul 02 16:13:47 <DiegoTc> jkridner: I'm worry about design. On my plans I though there will be a mockup for today. But there's not assuming worst case, do you feel comfortable the way it looks from 1 to 10? 10 being awsome
Jul 02 16:13:59 <vmayoral> I actually want to doxument it but it takes a while (big code 600k lines)
Jul 02 16:14:14 <jkridner> DiegoTc: 3. :-)
Jul 02 16:14:26 <vmayoral> Abhishek_: EKF
Jul 02 16:14:32 <Abhishek_> I see
Jul 02 16:14:49 <alexanderhiam> ds2: thanks, we hadn't thought of that. rseethamraju: so we may need to explicitely convert it from yuv to rgb or something like that
Jul 02 16:14:55 <vmayoral> And also DCMs but i'd need to check to be sure
Jul 02 16:15:06 <Abhishek_> k
Jul 02 16:15:12 <jkridner> DiegoTc: it'll be OK.... having the functionality there is a big thing and we can iterate on the experience a bit. Not as linear as is desired/efficient, but I think you are making something that will be usable with a bit of cleanup.
Jul 02 16:15:23 <rseethamraju> alexanderhiam: ya I’m googgling that
Jul 02 16:15:58 <jkridner> VoltVisionSteve: really want to know if help is coming and if you agree it is good to move on while waiting.
Jul 02 16:16:32 * VoltVisionSteve feels bad that the "help" was an email of advice.
Jul 02 16:16:38 <DiegoTc> jkridner: I will be here after the project ends, so I will continue working on it
Jul 02 16:16:58 <DiegoTc> if you agree of course
Jul 02 16:17:58 <VoltVisionSteve> jkridner, yes!  it is good to move on while waiting....we cant block.  Meanwhile I will bring the Guatemalan peeps into our next meeting so we can cut throug this.
Jul 02 16:18:40 <Abhishek_> karki_: which tool did you use to build that pdf?
Jul 02 16:18:49 <alexanderhiam> DiegoTc, jkridner, VoltVisionSteve: what parts of the design are you looking for help on? Just the layout in general, or is there a list of specific things that need design help?
Jul 02 16:18:51 <vmayoral|pc> jkridner: apologies (unreliable connection)
Jul 02 16:19:04 <karki_> Abhishek_ : Lucid charts
Jul 02 16:19:13 <karki_> It's amazing :)
Jul 02 16:19:14 <jkridner> alexanderhiam: general layout
Jul 02 16:19:21 <Abhishek_> karki_: indeed :)
Jul 02 16:19:33 <VoltVisionSteve> alexanderhiam, Diego is good on the JS/HTML, but we need help with the graphic design "prettiness".
Jul 02 16:19:34 <mranostay> morning
Jul 02 16:19:45 <alexanderhiam> right
Jul 02 16:19:58 <Abhishek_> hi mranostay, sigrok-cli seems to be a little slow for large sample sizes.
Jul 02 16:20:03 <VoltVisionSteve> alexanderhiam, the tool has to look "great" for people to use it and be enthused by it.
Jul 02 16:20:22 <jkridner> praveendath92: do you have thoughts on trying to find the failures?
Jul 02 16:20:24 <VoltVisionSteve> right now we have some prelim graphic design, but it needs that kick.'
Jul 02 16:20:29 <mranostay> Abhishek_: not threaded
Jul 02 16:20:51 <mranostay> ask biot about that?
Jul 02 16:20:56 <Abhishek_> yeah, it's an event loop
Jul 02 16:21:12 <Abhishek_> mranostay: see #sigrok, we chatted this evening about it
Jul 02 16:21:24 <jkridner> praveendath92: is debugging ADB+ADK required to get things done?
Jul 02 16:21:26 <Abhishek_> mranostay: ping me if you need the logs
Jul 02 16:21:42 <praveendath92> It's working now
Jul 02 16:21:42 <DiegoTc> VoltVisionSteve: Oscar answer right now. Not a positive answer
Jul 02 16:21:51 <praveendath92> No problem with that.
Jul 02 16:22:11 <praveendath92> I'm running some tests with XServer and framebuffer.
Jul 02 16:23:59 <jkridner> vmayoral: everything on schedule without blocking issues for beaglepilot?
Jul 02 16:24:44 <praveendath92> jkridner: what failures were you referring to?
Jul 02 16:24:53 <VoltVisionSteve> DiegoTc, so Oscar is saying for you to talk with Paola directly.  I feel terrible that this is potentially blocking your progress.  The free help isnt working out so well.
Jul 02 16:25:16 <vmayoral> jkridner: all good. After some discussions regarding the ROS integration I listened to the community and dived into the already available packages. I documented it http://erlerobotics.gitbooks.io/erlerobot/en/mavlink/mavlinkros.html
Jul 02 16:25:30 <vmayoral> jkridner: particularly promising http://erlerobotics.gitbooks.io/erlerobot/en/mavlink/ros/mavros.html
Jul 02 16:25:47 <DiegoTc> VoltVisionSteve: I'm guessing it will not have a positve effect. If Oscar that is his friend couldn't make her move, I don't I will
Jul 02 16:25:59 <DiegoTc> I don't think I will
Jul 02 16:26:11 <DiegoTc> I will tried and let you know
Jul 02 16:26:15 <VoltVisionSteve> alexanderhiam, I was trying to have a Guatemalan team help with the Graphic design for diego, but it is being challenging.  Do you have graphic design skills.?
Jul 02 16:26:50 <alexanderhiam> VoltVisionSteve: that depends on what you mean by graphic design skills...
Jul 02 16:27:09 <VoltVisionSteve> making the bone101 page look pretty...and entice people to use it!!
Jul 02 16:27:33 <jkridner> vmayoral: any thoughts on pushing Debian packages upstream?
Jul 02 16:27:39 <vmayoral> jkridner: The coming weeks are ROS, interface (Cloud 9 already available and i might take a look at Codiad) and Security. Sammy who was the mentor engaged for security purposes hasn't been responsive so the time could perfectly be allocated to push other aspects (out of the scope but as well interesting)
Jul 02 16:28:43 <vmayoral> jkridner: i contacted the package developer and pointed it out.
Jul 02 16:28:43 <jkridner> vmayoral: having your .deb packages be at least in debian.beagleboard.org would be good, but ideally you should target Debian 8 upstream.
Jul 02 16:29:00 <vmayoral> jkridner: sure, no problem from my side
Jul 02 16:29:38 <jkridner> entire install process on Beagle should be something like 'apt-get install beaglepilot' (or beaglepilot-XXX if choosing alternate stack)
Jul 02 16:29:56 <alexanderhiam> DiegoTc: how would I go about deploying your site? is the latest on github? I'm no graphic designer, but I'd be happy to take a look and see if I can help
Jul 02 16:29:58 <vvu> praveendath92: if you encounter any issues with fb stuff ask here, those X things can be rather hacky sometimes
Jul 02 16:29:59 <vmayoral> jkridner: can you share the steps needed to send it upstream (never done it before). A link would do it. For debian.beagleboard.org who should i get in touch with?
Jul 02 16:30:06 <VoltVisionSteve> all: I have to be across town in 1minute...sorry about the graphic design fiasco.  I can keep pushing on them, is that best?  I will bring them into our next meeting to come to terms with it.  Talksoon and sorry about it.
Jul 02 16:30:17 <vmayoral> jkridner: like a lot the "apt-get install beaglepilot".
Jul 02 16:30:22 <jkridner> vmayoral: post to the beagleboard google group and cc me. Rob and Robert should likely reply.
Jul 02 16:30:32 <DiegoTc> alexanderhiam: go ahead, need the link?
Jul 02 16:30:38 <alexanderhiam> sure
Jul 02 16:30:58 <praveendath92> vvu: Like you said I was trying to start xserver on fb.
Jul 02 16:31:01 <DiegoTc> alexanderhiam: https://github.com/DiegoTc/bone101/tree/nodejs
Jul 02 16:31:06 <ds2> bitbake beaglepilot
Jul 02 16:31:07 <ds2> :P
Jul 02 16:31:11 <cdsteinkuehler> Gotta head out...
Jul 02 16:31:41 <praveendath92> But then when I tried normally i got some success but after restarting my pc went into infinite loop.
Jul 02 16:31:44 <karki_> jkridner : I have posted my weekly update on Google groups. I have no current blockage from my side. I had all the issues sorted out thanks to the two hangouts :)  As for now I just have to code [I believe I know what to do and how to do ]
Jul 02 16:31:49 <jkridner> Ready to check in with Abhishek_?
Jul 02 16:31:56 <Abhishek_> here I am
Jul 02 16:32:05 <vvu> praveendath92: tried normally ? can you explain ?
Jul 02 16:32:13 <praveendath92> I fixed it. Now I will try it on the fbx and see what happens.
Jul 02 16:32:21 <vvu> okok perfect
Jul 02 16:32:22 <jkridner> Abhishek_: how much of a blocker is the speed issue? how unhappy about it are your mentors?
Jul 02 16:32:36 <praveendath92> vvu: i mean by just using startx with no destination specified.
Jul 02 16:32:42 <vvu> praveendath92: remember not do a lot of processing on the android
Jul 02 16:32:49 <praveendath92> It used my laptop screen.
Jul 02 16:32:53 <vvu> try there to keep just a main buffer on which you work
Jul 02 16:33:16 <Abhishek_> I think the block is mostly related to sigrok-cli and I would be developing something optimized to the BBB to work with libsigrok
Jul 02 16:33:21 <praveendath92> vvu: sure. I will keep that in mind.
Jul 02 16:33:34 <Abhishek_> which would be the backend to the web interface
Jul 02 16:33:48 <vvu> try maybe we can find some open source libs for image processing for android
Jul 02 16:33:52 <ds2> 6
Jul 02 16:33:56 <ds2> blah
Jul 02 16:34:17 <DiegoTc> jkridner: I assume you finish with bone101 some minutes ago>?
Jul 02 16:34:19 <vvu> i remmeber handling pics in apps on android can be hard...you run out of mem really fast. did that at a hackathon and spend some good hours to debug
Jul 02 16:34:56 <ds2> Abhishek_: I am not overly concerned... more interested to see how limiting is a 1M or so sized data set
Jul 02 16:35:04 <praveendath92> vvu: yes. I had issues with memory management with images on Android.
Jul 02 16:35:07 <jkridner> DiegoTc: was letting discussion continue with Steve, but I think you have clear actions for now, no?
Jul 02 16:35:16 <DiegoTc> yes
Jul 02 16:35:21 <DiegoTc> meeting this friday?
Jul 02 16:35:44 <vvu> praveendath92: i think the best way is to poll the resolution of the android device and create custom edid with it and send it to fb driver
Jul 02 16:35:46 <praveendath92> Let's see if my past experience is going to be of some help here.
Jul 02 16:35:52 <jkridner> DiegoTc: yes. do you know what time I need to schedule it? early, early again?
Jul 02 16:35:54 <DiegoTc> or you take of because it's 4 of July?
Jul 02 16:36:03 <praveendath92> vvu: I agree.
Jul 02 16:36:04 <vvu> devices with bigger resolution => bigger processing power
Jul 02 16:36:06 <jkridner> DiegoTc: I'm free in the morning.
Jul 02 16:36:08 <vvu> and so on
Jul 02 16:36:17 <praveendath92> The default is 1200x1024
Jul 02 16:36:26 <DiegoTc> 11AM your hour?
Jul 02 16:36:42 * jkridner is surprised by mention of July 4th.
Jul 02 16:36:58 <jkridner> DiegoTc: a bit earlier is better, but that can be fine.
Jul 02 16:37:03 <jkridner> 6AM wasn't so good. :-)
Jul 02 16:37:29 <DiegoTc> jkridner: 5Am? roosters will be singing
Jul 02 16:37:39 <jkridner> Abhishek_: I agree with ds2. 1MB is a more reasonable sample size.
Jul 02 16:37:42 <vvu> praveendath92: do you think until the end of the week we can have pixels on the android side ?
Jul 02 16:38:01 <Abhishek_> jkridner: I'll report some testing with software triggers in libsigrok.
Jul 02 16:38:12 <jkridner> DiegoTc: after 7AM US Eastern is desired, but I can do whatever before noon US Eastern.
Jul 02 16:38:19 <praveendath92> vvu: I think it is possible
Jul 02 16:38:25 <vvu> praveendath92: perfect if so
Jul 02 16:38:33 <vvu> if you need me to test smth just ping
Jul 02 16:38:42 <DiegoTc> jkridner: I willsend email to you and Steve to confirm
Jul 02 16:38:42 <vvu> also remember to document your stuff
Jul 02 16:39:04 <praveendath92> Sure. Document as I work?
Jul 02 16:39:22 <jkridner> Abhishek_: so, just a bit more performance testing and you'll be moving to the web interface stuff? or is there more that you'll do first?
Jul 02 16:39:26 <vvu> yep...you forget afterwards
Jul 02 16:39:32 * vvu speaking from experience 
Jul 02 16:40:15 <Abhishek_> jkridner: I have to tweak sigrok-cli to output something which the WaveDrom library should like and can be passed on to it for rendering. So the speed could be taken care of there
Jul 02 16:40:29 <praveendath92> vvu: I think so too. A lot of adk related issues are forgotten.
Jul 02 16:40:46 <vvu> yep true
Jul 02 16:41:00 <Abhishek_> jkridner: and soft triggers in the sigrok bindings, it is not too much of code.
Jul 02 16:41:18 <praveendath92> Reports could help recollect some but I will maintain a rough document.
Jul 02 16:41:18 <jkridner> Abhishek_: so, you're moving on to the web stuff now with only minor (non time killing) distractions?
Jul 02 16:41:42 <karki_> jkridner : What about the BBB for me and Abhishek_ ?
Jul 02 16:41:46 <Abhishek_> jkridner: I think by next week, I would be working on the web stuff
Jul 02 16:42:05 <Abhishek_> after clearing the patch for sigrok
Jul 02 16:42:11 <jkridner> karki_, Abhishek_: I'm totally throttled by TI India as they are handling the import. :(
Jul 02 16:42:27 <karki_> :'(
Jul 02 16:42:33 <jkridner> it seems all of the distributors haven't taken in any orders recently and all are out of stock.
Jul 02 16:42:38 <jkridner> (in India)
Jul 02 16:42:46 <vvu> seems that we have nice going projects this year :)
Jul 02 16:42:52 <Abhishek_> jkridner: No need to redirect my shipment now, please ensure it reaches me when I reach India on 15th.
Jul 02 16:43:13 <jkridner> Abhishek_: yeah, I'd already given up on catching you out of the country.
Jul 02 16:43:28 <Abhishek_> :)
Jul 02 16:43:43 <jkridner> hardware is always a pain, but it is also what makes things fun.
Jul 02 16:44:01 <jkridner> rseethamraju: did you have any more items to cover on PyBBIO?
Jul 02 16:44:08 <vmayoral> indeed (both the pain and fun stuff :) )
Jul 02 16:44:11 <Abhishek_> jkridner: indeed, I am more of a hardware guy so I know the fun quite well :)
Jul 02 16:44:23 <rseethamraju> jkridner: no
Jul 02 16:44:46 <ds2> vvu: as long as they all finish in time :D
Jul 02 16:45:08 <vvu> +1
Jul 02 16:45:36 * vvu 's mutt went crazy
Jul 02 16:46:44 <jkridner> karki_: I know you said you had no issues right now, but how are you feeling about the eventual moving of the python tokenizer code into the kernel?
Jul 02 16:47:16 <Abhishek_> my time's going to be limited after 16th July due to university classes but I'll work it out.
Jul 02 16:47:31 <vvu> so soon Abhishek_ ?
Jul 02 16:47:46 <rseethamraju> Abhishek_: mine already started!
Jul 02 16:47:47 <jkridner> karki_: personally, I feel like moving the tokenizer into the kernel will be easy and as long as you can handle the SCRIPT/ENDSCRIPT with your approach now.
Jul 02 16:47:48 <rseethamraju> =(
Jul 02 16:47:58 <jkridner> rseethamraju: ?
Jul 02 16:48:06 <karki_> jkridner : Thats not an issue for this week or the next ;) First I want to get everything working and tested by Chris! Then we can make it better/efficient!
Jul 02 16:48:17 <rseethamraju> oh nothing college started
Jul 02 16:48:20 <karki_> i.e. moving stuff to kernel space
Jul 02 16:48:35 <Abhishek_> vvu: yup, I suppose me and praveendath92 had our vacations started quite early, in about a week since acceptance.
Jul 02 16:48:42 <jkridner> rseethamraju: lack of good advice on gstreamer w/ webcams?
Jul 02 16:49:15 <praveendath92> Happy that I'm just done with college :)
Jul 02 16:49:28 <karki_> +1
Jul 02 16:49:29 <praveendath92> Nothing till October :)
Jul 02 16:49:30 <jkridner> karki_: that's fine. did you ever sit down and write the 1-to-1 relationship between script and numerical tokens I requested?
Jul 02 16:50:18 <rseethamraju> jkridner: no I got it. We didn’t think of color space being the issue. We’ll try that tomorrow. I’m working on the openCV alterative anyway. Incase we can’t get that to work.
Jul 02 16:50:20 <jkridner> karki_: I still don't think you know the worst-case instruction size. I like having a fixed instruction size for reasons you've mentioned in the past, but I think you need that design document and you are overlooking things.
Jul 02 16:50:34 <karki_> jkridner : I'm on it :) I've been doing it bit by bit last week. It should be done by this week, along with the implementation.
Jul 02 16:50:46 <karki_> jkridner : don't worry :)
Jul 02 16:50:53 <jkridner> rseethamraju: k. I've been playing with python-cv quite a bit (http://github.com/jadonk/Dish-Detector)
Jul 02 16:51:01 <karki_> I did over look some stuff initially
Jul 02 16:51:03 <karki_> but
Jul 02 16:51:11 <jkridner> karki_: famous last words. ;-)
Jul 02 16:51:14 <karki_> now I feel I know how to go about it!
Jul 02 16:51:18 <karki_> ;)
Jul 02 16:51:40 <jkridner> did I miss anybody or any issues?
Jul 02 16:51:45 <alexanderhiam> karki_: have you figured out your maximum number of user variables yet?
Jul 02 16:52:19 <vmayoral> jkridner: are you in good terms with the OpenROV guys? What do you think about asking them to join the ardupilot efforts enhacing the autopilot for underwater vehicles as well?
Jul 02 16:52:41 <vmayoral> jkridner: if you get in me in touch with them i can do the pitch
Jul 02 16:52:44 <jkridner> vmayoral: interesting.
Jul 02 16:53:06 <jkridner> vmayoral: I suggest you join their G+ hangout.... which is what they've asked me to do in the past but it never ends up on my calendar.
Jul 02 16:53:20 <jkridner> vmayoral: I'll foster an introduction.
Jul 02 16:53:47 <vmayoral> jkridner: thanks!
Jul 02 16:54:10 <karki_> alexanderhiam : did you see my discussion with panto ? Google hangouts one. we decided that we will store some space on the 8KB ram. I'm looking at a max of 256 variables. (as of now)
Jul 02 16:55:01 <alexanderhiam> karki_: yea h I watched that, did I miss you saying 256 in it?
Jul 02 16:55:17 <karki_> I didn't mention anything then
Jul 02 16:55:22 <alexanderhiam> ok
Jul 02 16:55:24 <karki_> but we need a max limit
Jul 02 16:55:43 <karki_> so as of now I thought 256 is a safe limit
Jul 02 16:55:59 * jkridner sounds gavel and encourages continued discussion
Jul 02 16:56:02 <jkridner> thanks all
Jul 02 16:56:11 * ds2 runs like mice on a sinking ship
Jul 02 16:56:12 <ds2> ;)
Jul 02 16:56:13 <alexanderhiam> karki_: yeah, it would be hard to use that many in a script
Jul 02 16:56:35 <alexanderhiam> karki_: I can't remember, does BotSpeak support arrays?
Jul 02 16:56:45 <karki_> yup
Jul 02 16:56:46 <jkridner> karki_, alexanderhiam: +1 256 variables seems like plenty for a BotSpeak script.
Jul 02 16:56:54 <jkridner> alexanderhiam: yeah, it does.
Jul 02 16:56:55 <rseethamraju> jkridner: nice! Thanks
Jul 02 16:58:52 <_av500_> back
Jul 02 16:58:56 <_av500_> just in time :)
Jul 02 16:59:03 <vvu> _av500_: exactly in time
Jul 02 17:00:08 <alexanderhiam> rseethamraju: this works with my webcam: https://github.com/jadonk/Dish-Detector/blob/master/camera-test.py (thanks for the link jkridner!)
Jul 02 17:00:24 <rseethamraju> cool
Jul 02 17:00:39 <alexanderhiam> rseethamraju: might be worth digging through the opencv source to see if you can figure out how they're setting up that pipline
Jul 02 17:00:49 <rseethamraju> I’ll call the video record func then
Jul 02 17:00:58 * vmayoral is goint to try to fly a copter
Jul 02 17:01:08 <vmayoral> I'll be around guys, ping me if anyone wishes anything
Jul 02 17:01:15 <Abhishek_> good luck vmayoral :)
Jul 02 17:01:29 <rseethamraju> alexanderhiam: theres an element to read from a file
Jul 02 17:01:59 <rseethamraju> filesrc
Jul 02 17:02:22 <alexanderhiam> right
Jul 02 17:02:46 <rseethamraju> but for streaming it has to capture frames and send
Jul 02 17:03:04 <rseethamraju> so I’ll look into how that’ll work
Jul 02 17:03:08 <alexanderhiam> ok
Jul 02 17:03:35 <rseethamraju> I mean we can’t wait for it to save to a file and then read and send it
Jul 02 17:04:46 <alexanderhiam> right, but I mean opencv clearly sets up a pipeline that works for this camera, so presumably that same setup up to the filesink could work connected to a tcpserversink
Jul 02 17:05:59 <alexanderhiam> DiegoTc: so I have node.js delivering pages from a clone of your repo, but I'm not seeing how to get to your stuff
Jul 02 17:06:11 <alexanderhiam> I'm just seeing the normal bone101 pages
Jul 02 17:08:35 <rseethamraju> can we try a color space element and see too?
Jul 02 17:08:48 <alexanderhiam> sure
Jul 02 17:09:15 <DiegoTc> alexanderhiam: https://github.com/DiegoTc/bone101/tree/nodejs/Support/GSOC/app/views
Jul 02 17:09:27 <Abhishek_> jkridner: did I tell you that testing the RNDIS link using iperf -s on the PC side, the BBB clocks 80 Mbps but if I do iperf -s on the BBB side, only 40 Mbps?
Jul 02 17:09:33 <DiegoTc> on Support-> GSOC -> App -> views
Jul 02 17:12:42 <rseethamraju> alexanderhiam: PyBBIO doing that thing again where it doen’t install the changes in spi module. This time I deleted dist/ and PyBBIO.egg.info/ any new suggestions?
Jul 02 17:13:12 <alexanderhiam> DiegoTc: 'Cannot GET /Support/bone101/GSOC/app/views'
Jul 02 17:14:03 <DiegoTc> alexanderhiam: sure you clone the nodejs branch?
Jul 02 17:14:03 <alexanderhiam> rseethamraju: build/?
Jul 02 17:14:14 <alexanderhiam> DiegoTc: oops
Jul 02 17:14:19 <rseethamraju> oh right sorry
Jul 02 17:15:36 <alexanderhiam> DiegoTc: thanks, got it
Jul 02 17:15:56 <DiegoTc> yeah, that happen to me the first time, I clone it on a different computer
Jul 02 17:17:05 <alexanderhiam> DiegoTc: fwiw on the nodejs branch the bone101 page doesn't connect to my bone and cloud9 doesn't load on port 3000
Jul 02 17:34:21 <rseethamraju> alexanderhiam: it says theora supports YCbCr
Jul 02 17:45:03 <jkridner> Abhishek_: didn't know that. reason?
Jul 02 17:45:43 <Abhishek_> I suppose it has something to do with the TCP overhead, I'm not very sure about that though.
Jul 02 19:34:43 <_av500_> jkridner: can you reply to stephanie?
Jul 02 19:36:04 <jkridner> _av500_: was going to ask about this reunion thing, but you were out of pocket.
Jul 02 19:36:38 <_av500_> I understood me and vvu are the "delegates"
Jul 02 19:37:03 <jkridner> right....  Just sucks that vvu "won the lottery" and we can't use
Jul 02 19:37:38 <_av500_> right
Jul 02 19:38:14 <Abhishek_> use?
Jul 02 19:40:22 <vvu> yep, they won't let me trade the spot
Jul 02 19:40:38 <vvu> it goes back into the game
Jul 02 19:41:20 <Abhishek_> oh
Jul 02 19:41:47 <vvu> i asked her it that is possible to assign it for bb but does not work that way
Jul 02 19:41:59 <_av500_> rulz are rulz
Jul 02 19:41:59 <Abhishek_> I see
Jul 02 19:42:29 <Abhishek_> however the 2 slots of the organization are independent of anyone winning the lottery IIRC
Jul 02 19:42:39 <vvu> yes
Jul 02 23:08:10 <Abhishek_> jkridner: soft triggers are working with sigrok and BeagleLogic now
Jul 02 23:46:38 <jkridner> cool!
Jul 02 23:47:32 <Abhishek_> However, there are no pre-trigger samples currently, I would add that soon
Jul 02 23:48:15 <Abhishek_> implementing that would need some tweaks in libsigrok
Jul 02 23:54:31 <Abhishek_> http://s28.postimg.org/ro409bw2l/sigrok_bbb_ascii.png
Jul 03 00:39:22 <Abhishek_> jkridner: http://thread.gmane.org/gmane.comp.debugging.sigrok.devel/1014
Jul 03 02:35:21 <jkridner> Abhishek_: awesome. You might want to put in the comments a home page to find the firmware.
Jul 03 02:35:51 <jkridner> And where to find info about the BeagleBone Black.
Jul 03 02:36:14 <jkridner> Odd that your first email in it doesn't have a subject
**** ENDING LOGGING AT Thu Jul 03 03:00:00 2014